RTFM

Install

Works with Claude Code, Codex, VS Code, Cursor, Windsurf, Claude Desktop. Read the repository before you give any tool access to your data.

npx agentmarket install connector/roomi-fields-rtfm

About

Use cases: knowledge-base, document-processing

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this server can do and what it asks for, from its tool list, install config and code. Checked Oct 25, 2026.

Tools

8 read · 5 write · 1 delete

Secrets

None asked for

Install scripts

None found

Runs as

Python package

Tools (14)

Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].

• rtfm_books

  List indexed books with per-corpus summary and pagination. Args: corpus: filter by corpus name (optional) limit: max books per page (default 50, 0 for all) offset: skip N books for pagination (default 0) project: another project's index to read instead of this one — its name (``"hub"``), or its path when two projects share a name.

• rtfm_context

  Get sources relevant to a subject. Returns paths + line ranges, no content. Use rtfm_expand to read. Args: subject: topic, concept, file path, or question scope: corpus filter (optional) limit: max sources (default 5) project: another project's index to read instead of this one — its name (``"hub"``), or its path when two projects share a name.

• rtfm_coverage

  How much of a project this index actually holds. Say this before answering from the index when completeness matters: a partial index read as a complete one is how an absent result becomes "there is nothing on the subject". The denominator is the scan's own list of files — not everything in the directory. Logs, lock files, state files and build output are not counted as gaps, because RTFM was never going to index them. Args: project: another project to measure — its name ("hub"), or its path when two projects share a name. Omit for the current one.

• rtfm_discover

  Scan a project directory and return a structural map (file types, languages, entry points, size breakdown). ~1 second. Args: path: project root directory (default ".")

• rtfm_expand

  Read content of an indexed file with line numbers. Use after rtfm_search. Like Read, but for indexed files. Args: source: absolute file path (from search results) target: jump to section name ("class Foo") or "L120" query: filter chunks by relevance within the file offset: pagination offset (default 0) count: chunks to return, 0 for all remaining (default 1) project: another project's index to read instead of this one — its name (``"hub"``), or its path when two projects share a name.

• rtfm_graph

  Show graph neighbors of a file (imports, links, includes, citations). Args: source: book slug or absolute file path direction: "outgoing" (deps), "incoming" (dependents), or "both" relation_type: filter by "import" | "link" | "include" | "cite"

• rtfm_history

  Show version history of an indexed file, or retrieve a specific version. Each time a file is re-synced, its previous content is saved as a snapshot. Args: source: file path (an identity also works). version: Version number to retrieve (optional). If omitted, lists all versions. project: another project's index to read instead of this one — its name ("hub"), or its path when two projects share a name.

• rtfm_ingest

  Ingest a single file into the library. Args: path: Path to the file to ingest. corpus: Corpus name (default: "default").

• rtfm_remove

  Remove a file and its chunks from the library. Args: filepath: The filepath (as tracked in the index) to remove. corpus: Which corpus it belongs to. A relative path identifies a file only within its corpus — the same README.md can be indexed in several — so removing without saying which one is ambiguous.

• rtfm_search

  Search the indexed knowledge base. Returns ranked source paths with line ranges, no content. Use rtfm_expand to read content. Args: query: search query limit: max sources (default 5) corpus: filter by corpus name search_type: "fts" | "semantic" | "hybrid" freshness_weight: boost recently indexed files (0.0–0.5) centrality_weight: boost files with many incoming edges (0.0–0.5) project: another project's index to search instead of this one — its name (``"hub"``), or its path when two projects share a name. Omit for the project you are working in.

• rtfm_stats

  Get library statistics: total chunks, books, corpora, tag and embedding coverage. Also reports the install state of optional extras (embeddings, pdf) and surfaces actionable recommendations when the current index would benefit from an extra the user has not installed.

• rtfm_sync

  Sync a directory into the library. Only processes changed files. Args: path: Directory to sync (default: current directory). corpus: Corpus name for indexed documents (default: "default"). extensions: Comma-separated file extensions to include (e.g. "md,py,pdf").

• rtfm_tag_chunks

  Add tags to specific chunks. Args: chunk_ids: Comma-separated chunk IDs to tag. tags: Comma-separated tags to add.

• rtfm_tags

  List all tags in the library, optionally filtered by corpus. Args: corpus: Filter by corpus name (optional).

Compare and alternatives

Featured in

• Best MCP servers for PDF

Signals

Stars

27

Forks

6

Open issues

4

Contributors

2

Weekly downloads

538

Last commit

Oct 22, 2026 (6 days ago)

Last release

Oct 22, 2026 (6 days ago)

Last package publish

Oct 22, 2026 (6 days ago)

From GitHub (fetched Oct 24, 2026) and the package registry.

Details

Registry name

io.github.roomi-fields/rtfm

Version

0.55.1

Auth

No auth

Packages

pypi rtfm-ai

License

MIT

Repository

github.com/roomi-fields/rtfm

Website

rtfm.roomi-fields.com/

First seen

Oct 24, 2026

Last checked

Oct 24, 2026

Source

Official MCP Registrygithub-graphql